Why preferential hydration does not always stabilize the native structure of globular proteins

The observed preferential hydration of proteins in aqueous MgCl2 solutions at low pH and low salt concentration (Arakawa et al., 1990) prompted a scrutiny of possible protein stabilization by MgCl2 under the same conditions, in view of earlier observations in aqueous solutions of sugars, amino acids, and a number of salts that preferential hydration is usually accompanied by the stabilization of the native structure of globular proteins. The results of thermal transition experiments on five proteins (ribonuclease A, lysozyme, beta-lactoglobulin, chymotrypsinogen, and bovine serum albumin) revealed neither significant stabilization nor destabilization of the protein structures by MgCl2 both at acid conditions (except for ribonuclease A, which was stabilized, but to a much smaller extent than by MgSO4) and at higher pH at which MgCl2 displayed little preferential hydration. This was in contrast to the great stabilizing action of MgSO4 at the same conditions. 2-Methyl-2,4-pentanediol (MPD), which gives a very large preferential hydration of native ribonuclease A at pH 5.8 Pittz & Timasheff (1978) Biochemistry 17, 615-623], was found to be a strong destabilizer of that protein at the same conditions. Analysis of the preferentially hydrating solvent systems led to their classification into two categories: those in which the preferential hydration is independent of solution conditions and those in which it varies with conditions. The first always stabilize protein structure, while the second do not. In the first category the predominant interaction is that of cosolvent exclusion, determined by solvent properties, with the protein being essentially inert. In the second category interactions are determined to a major extent by the chemical nature of the protein surface.(ABSTRACT TRUNCATED AT 250 WORDS)